It is proposed to exploit the unique ethnic origin mix of Jerusalem to measure the prevalence of the senile dementias in a representative population sample over 65 and the effects of age, sex, origin, civil status, education, living conditions, family and social support, religiosity and patterns of morbidity on frequency and severity of disease. This will be done by a two-stage process, screening of over 4,000 individuals followed by detailed assessment and medical examination of the 10% or so who fail the screening test. Matched controls will be interviewed for ADL and social factor comparisons. The first phase will be the development and testing of the screening and assessment instruments which will be designed to be as culture free as possible and provide weighted scores correlating highly with the final diagnosis. They will be designed to be administered by partially trained interviewers and suitable for mass screening as well as repeated monitoring of cases. Subsequent studies are planned to measure incidence and the natural history of the different dementias under various social, cultural and environmental conditions, to develop risk scores for early diagnosis and intervention and to identify sick populations for detailed biomedical studies.